– DRAFT –
DXWG Plenary

<PWinstanley> egrets+ annette

Admin

PWinstanley: any additions to agenda?

<PWinstanley> https://‌www.w3.org/‌2019/‌10/‌15-dxwg-minutes

proposed: accept minutes of Oct 15

<roba> +0

<ncar> +1

<riccardoAlbertoni> 0 ( i was not there)

<DaveBrowning> +1

+1

Resolved: accept minutes of Oct 15

Open Actions

PWinstanley: look at them later, if time

DCAT CR

DaveBrowning: I've been out so I have nothing to report

Andrea: (not on IRC yet) All links on agenda.

<PWinstanley> Draft implementation report: https://‌raw.githack.com/‌w3c/‌dxwg/‌andrea-perego-dcat-implementation-report/‌dcat-implementation-report/‌index.html

Andrea: there's a draft implementation report on github

<PWinstanley> Statistics on collected implementations: https://‌docs.google.com/‌spreadsheets/‌d/‌1eEVUuPFAGO2GjS5ocxylY8T1AlpqlwnOTc3er_Mhcv4/‌edit#gid=1350045555

Andrea: a few things to check regarding charter; seems we only need one implementation for each feature update
… most new terms have one or more than one
… this seems to be the threshhold we need to meet

PWinstanley: charter says 'at least one implemention' - is that enough?

dsr: we'll need to defer to philippe - it's possible that the rules are different for vocabularies
… for software there are higher standards

Action: PWinstanley to ask PLH about number of implementations

<alejandra> the question is about the implementation of a single property, right?

<trackbot> Created ACTION-380 - Ask plh about number of implementations [on Peter Winstanley - due 2019-10-29].

PWinstanley: we have a lot of evidence in major efforts that the standard is being accepted

dsr: maybe we could change the wording

PWinstanley: Andrea's document provides a lot of evidence, and there is a lot of commitment from the community

andrea: we also have gathered planned implementations
… plus there is work to align DCAT-AP to align with DCATv2
… DCAT-AP is being extended at the national level for those APs
… so at least one or more extensions of DCAT-AP will be implemented
… some are waiting for DCATv2 to become a recommendation
… so are planned implementations counted as implementation evidence?

dsr: using actual implementation precedes CR finalization

AndreaPerego: the DCAT-APs will not be aligned until DCATv2 is finalized
… most changes reflect requests of the DCAT-A
… reflect requests that came from DCAT-APP extensions
… looking at existing implements we found at least one implementation; for the 24 that are not yet implemented there is at least one planned one

AndreaPerego: deadline is the end of this month, right? (answer: yes)
… so we need to know ASAP if this is sufficient

PWinstanley: absolutely! on it!

riccardoAlbertoni: maybe we should wait for the answer from plh; could there be an extension? or does CR wait?

PWinstanley: you're offering some other implementations, right?

ncar: I have some implementations of DCAT in the Conneg work; also the CKAN extension, which will implement nearly all properties
… I'll send to Andrea later today

<AndreaPerego> Thanks, ncar

PWinstanley: let's try to finish this this week, over email etc.

Conneg

ncar: met last week and addressed action items from the track and we've closed most of them
… may only have one left
… didn't substantially address issues in call last week

<ncar> Conneg by P implementation: https://‌conneg.info/‌mediatypes/‌connegp

ncar: big news item is there is one implementation of all aspects of the specification
… that link is to a service that shows an implementation; and collects the normative requirements (under red boxes)

<roba> no substantive issues in play - processing some editorial cleanups and double checking all actions addressed.

ncar: and shows how they have been implemented

<roba> DCAT usage examples highlighted - wanted feedback from DCAT team this is correct

ncar: moving on to a second implementation of some of the features for the CKAN plugin
… that gives us both DCATv2 and Conneg implementations
… roba has another implementation for OGC

roba: there was one other example we wanted to include describing how DCAT references Conneg profiles

<ncar> The Media Types Data Service dCAT2 implementation: https://‌conneg.info/‌mediatypes?_format=text/‌turtle

roba: issue #1042

<roba> https://‌github.com/‌w3c/‌dxwg/‌issues/‌1042

PWinstanley: next point we have to address: conneg is likely to be within this charter period will be a note
… and we need to complete that by the end of 2019
… and when we recharter it has time to be on rec track as CR
… need to secure it as a working group note, with all of what we've been talking about here in place
… so when new charter is agreed this goes easily into rec track and can move into CR smoothly
… this is a management issue that we need to keep in mind
… need a well-written note by end of charter period

ncar: I think that is in line which what we have been doing
… and helps support DCAT as well

ncar: if example is wrong then that's an issue for DCAT

<PWinstanley> kcoyle: to ask if anyone knows the status of the IETF document

<PWinstanley> ncar: the deadline has been extended

ncar: has been extended - was end of 2019

<ncar> the IETF ConnegP doc: https://‌profilenegotiation.github.io/‌I-D-Profile-Negotiation/‌I-D-Profile-Negotiation

<ncar> Expires: March 13, 2020

ncar: Conneg and IETF are in sync

PWinstanley: do we have implementation on http as well as query strings?

ncar: yes

roba: yes

ncar: implements all four profiles
… nothing is required for IETF

roba: I need the QSA or cannot use it; so now I have implementations

Re-Charter

PWinstanley: draft is circulating https://‌raw.githack.com/‌w3c/‌dxwg/‌recharter_2/‌charter/
… also there's a poll
… raised in the DCAT meeting and poll that evergreen is not yet available; still doing "waterfall" process
… this is delicate because standards bodies get scrutiny for collusive behavior
… there needs to be clarity about the scope relating to UCR and wide review to a larger population
… need to show that it's not just a small group of people serving mainly themselves
… W3C is therefore strict about IPR and review steps
… so for evergreen there needs to be working out that maintains this clarity
… we've been advised to have a one-year charter period, during which we will transition to the evergreen model as it is worked out by W3C
… that's the answer to the comments about the "old style" charter
… we don't need to create a full UCR document, but we do need to bring over the ones that are relevant to the features that we'll work on during that year
… but we also need to make sure that we are heavily involved with community organizations

<Zakim> DaveBrowning, you wanted to ask what could be done in a notional 1 year window...

PWinstanley: please move speedily with feedback so we can get the draft done by next week

DaveBrowning: I agree with the intent; but specific point is that if we have a one year window and the external bodies slow things down because they are busy
… a one year window makes it hard to imagine that we'll get something done

PWinstanley: how about a 3 year charter?

DaveBrowning: that sends a certain message
… if we say one year then we have to lock things down by 4 months in
… it seems like short window

PWinstanley: we need to develop a pipeline that allows us to bite off small amounts of change and move those through

PWinstanley: but there are many who are champing away at the question of collusion, which is why we need the widest review

PWinstanley: should i try to negotiate a 2-year charter?

DaveBrowning: my personal view is I'd be interested to see what comments that elicits

PWinstanley: there's going to be a lot of liaison for this to work smoothly

alejandra: I understand the constraints; there is a possible compromise; we probably don't need to create a new UCR document
… that doesn't help us progress; so instead of a new UCR, use github milestones, without having to create a new UCR document
… but people could add new use cases

<DaveBrowning> +1 to building the work off the current ucr (+/- anything new and important coming in)

PWinstanley: we would bring across from old UCR things that are relevant to the new project

alejandra: working on a UCR document itself is not a good use of our time because the use cases are already in the old document
… but we don't spend time on a new UCR document

PWinstanley: my view is that we can be flexible; I'll change the wording to show that we will not wait on a UCR document before starting
… work and we'll bring things over; and we'll use github issues which are already marked for future implementation

alejandra: yes, please make those clarifications

roba: there are not outstanding issues for conneg;

PWinstanley: this all happens in parallel
… need to engage from the start with those who will be involved in review

roba: conneg would be ok with the shorter timeline

PWinstanley: we aren't starting from scratch, but moving from here forward, bringing things over
… may separate out some things in github
… the charter has to be the #1 step or we have no future plan

<PWinstanley> https://‌www.w3.org/‌2017/‌dxwg/‌track/‌actions/‌open

Open Actions

(many listed as closed)

PWinstanley: please respond to poll

<PWinstanley> poll - https://‌www.w3.org/‌2002/‌09/‌wbs/‌99375/‌charter_2

<PWinstanley> https://‌raw.githack.com/‌w3c/‌dxwg/‌recharter_2/‌charter/

<PWinstanley> one final thiing I need to mention

<alejandra> thanks, and bye!

<PWinstanley> is that Caroline Burle is likely to join as co-chair for next session

<PWinstanley> bye

<AndreaPerego> So, welcome back, Caroline.

<alejandra> +1